What’s so weak about “weak-form review”? The case of the UK Human Rights Act 1998

Abstract

1. Introduction

In recent comparative constitutional law, there has been a surge of interest in the constitutional developments which have occurred in Canada, New Zealand, and the United Kingdom. ¹ Many scholars have grouped these countries together, arguing that their bills of rights instantiate a “new model of constitutionalism” which has been variously described as “the new Commonwealth model,” ² “the parliamentary model,” ³ the “hybrid approach,” ⁴ or “weak-form review.” ⁵ No matter which label is used, all scholars agree that the distinctive feature of these bills of rights is that while they enhance the powers of the courts to protect rights, they nonetheless allow the elected branches of government to have a greater input into decisions about rights. ⁶ As Stephen Gardbaum has put it, the new model of constitutionalism “decouples judicial review from judicial supremacy by empowering the legislature to have the final word.” ⁷ This “de-coupling” is achieved in various ways. For example, while the Canadian Charter of Rights and Freedoms gives the courts the power to strike down legislation which violates Charter rights, § 33 enables the legislature to legislate “notwithstanding” those rights. ⁸ In the UK, the courts are empowered to make a declaration that legislation violates rights, but this declaration has no impact on the validity or effect of the legislation under scrutiny and places no legal obligation on Parliament to change the law. ⁹ Both of these mechanisms allow for some form of legislative dissent from judicial rulings. ¹⁰ They seem to give the legislature (rather than the courts) the last word on questions about rights.

The focus of this article will be on one single jurisdiction (the United Kingdom) and one way of conceptualizing that system, namely, the idea suggested by Mark Tushnet that it is an instance of “weak-form” rather than “strong-form review.” ¹¹ Mark Tushnet’s distinction has become an enormously influential way of characterizing, and then distinguishing between, the US system of rights-protection on the one hand, and that which exists in Canada, the UK, and New Zealand on the other. ¹² The typology is now used to characterize many other systems ¹³ and has become a commonplace descriptor in comparative constitutional law scholarship. The aim of this article is threefold. The first is to explicate the legal and political dynamics surrounding the operation of the UK Human Rights Act (HRA). The second is to show that if we look at the UK system through the lens of the distinction between strong-form and weak-form review, we risk distorting and misunderstanding some of the key features of that system. The third is to put the distinction between strong-form and weak-form review under a probing analytical spotlight.

I will identify three main limitations of the distinction, all of which are interrelated. First, it focuses too narrowly on the formal features of the constitutional design, detached from how those features operate in practice. Second, it cashes out the metric of strength using the crude idea of institutional “displacement” (and the related idea of which institution has the “last word” on rights questions), thus obscuring and underplaying the important role of other devices and techniques which judges use in constitutional judicial review. Finally, I will argue that there is an inherent instability in the way the distinction is made out, because Tushnet uses it both as a typology of different systems of review and to refer to various devices and techniques which are shared across strong-form and weak-form systems alike. I will argue that this undercuts the usefulness of the distinction as a meaningful typological tool for comparative constitutional law and reveals tensions in Tushnet’s approach to classification of different systems.

The article will proceed in the following way. Section 2 will sketch out the central features of Tushnet’s distinction between strong-form and weak-form review. Section 3 will provide an overview of the system of rights-protection under the UK Human Rights Act 1998. Section 4 will examine the way in which Tushnet applies the distinction to the UK system. Finally, Section 5 contains a critical analysis of the distinction itself, highlighting some of its limitations and instabilities both as a typological tool for comparative constitutional law and as a way of understanding the nature of constitutional judicial review more generally.

2. The strong-form/weak-form distinction examined

Mark Tushnet introduced the distinction between strong-form and weak-form review as a way of illuminating the contrast between US-style judicial review on the one hand, and the system of rights-protection in the UK, Canada, and New Zealand on the other. ¹⁴ The nomenclature of the distinction suggests that we should distinguish between these systems along the dual dimensions of strength and form . How are these two dimensions specified? This question is harder to answer than one might think, because there is some shift of emphasis across Tushnet’s voluminous writings on the subject. ¹⁵ Nonetheless, the central elements which go to make up the “basic distinction” ¹⁶ are relatively clear.

According to Tushnet, “strong-form review is a system in which judicial interpretations of the Constitution are final and unrevisable by ordinary legislative majorities.” ¹⁷ In many of his writings, Tushnet treats the US system as an archetype or paradigm of strong-form review. ¹⁸ Thus, he states:

In systems with strong-form judicial review, of which the United States is usually taken to be representative, the constitutional court has the power to invalidate primary legislation, and legislative responses to such invalidations are made quite difficult. ¹⁹

Tushnet suggests that there might be different variants of strong-form review, ²⁰ but argues that there is one key feature which unites all the variants, namely, that the courts have “normative finality” over the legislature on questions of rights. ²¹ Under strong-form review, the courts can “displace” ²² legislative decisions about rights, in a situation where that displacement is “conclusive” or “final.” ²³ As he puts it: “strong-form review insists that the courts’ reasonable constitutional interpretations prevail over the legislatures’ reasonable ones.” ²⁴

By contrast, the central characteristic of weak-form review is that “courts assess legislation against constitutional norms, but do not have the final word on whether statutes comply with those norms.” ²⁵ As he puts it:

Weak-form systems of judicial review provide an opportunity for judicial oversight of legislation without displacing the ultimate power of legislatures to determine public policy. ²⁶

So, we might say that whereas under strong-form review, the courts can displace the legislature (in a way which is effectively final and unrevisable), “the mark of weak-form review is that ordinary legislative majorities can displace judicial interpretations of the constitution in the relatively short run.” ²⁷ Rather than being final or conclusive or even necessarily binding on the other branches, court rulings under weak-form review are “expressly open to legislative revision in the short run” ²⁸ and are therefore “provisional.” ²⁹

As with strong-form review, Tushnet suggests that weak-form review can be instantiated in a number of different ways. ³⁰ In Tushnet’s view, the weakest variant is an interpretive mandate which allows the courts to interpret legislation compatibly with constitutional norms if at all possible. ³¹ Another (stronger) version is a judicial power to “declare legislation to be inconsistent with constitutional norms, but not to enforce such judgments coercively against a losing party.” ³² Still another is that which obtains in Canada, namely, a system which allows the courts to enforce its judgment coercively (by invalidating the legislation or relying on remedies such as mandatory injunctions or damages), but allows the legislature to “respond by reinstating the original legislation by some means other than a cumbersome amendment process.” ³³ He concludes:

Specific forms vary, but all share the characteristic that the court decision ultimately resolving a constitutional matter can be followed by a legislative or popular response short of constitutional amendment that differs in whole or in part with the courts’ decision. ³⁴

By characterizing weak-form review as “provisional” and subject to legislative response or override, it might be tempting to associate Tushnet with the view that weak-form review has a distinctively “dialogic” quality which is absent from strong-form review. ³⁵ It is certainly true that Tushnet refers on occasion to the “iterative structure of weak-form review” ³⁶ and he highlights the way in which “weak-form review purports to promote a real-time dialogue between courts and legislatures.” ³⁷ However, he is careful not to claim that a dialogic process of exchange between courts and legislatures is the exclusive property of weak-form review. In fact, he acknowledges that even in strong-form systems, some legislative response to court decisions is possible although it may to be more difficult to achieve. ³⁸ Thus, for Tushnet, the difference between weak- and strong-form review does not map onto a bright-line distinction between dialogic and non-dialogic review. Rather, it resides in “the relative ease or difficulty of the legislative response to judicial determinations” ³⁹ and the time-frame within which such dialogue can occur. ⁴⁰

When Tushnet first introduced the distinction between strong-form and weak-form review, it was pitched at the level of institutional and constitutional design. ⁴¹ Thus, he portrayed strong- and weak-form review as embodying different models of constitutionalism, ⁴² which differed or varied depending on how judicial review was structured or institutionalized in different legal systems. ⁴³ Therefore, the “form” in “strong-form review” refers to the way in which the powers of the courts and legislatures are designed with respect to rights-protection. ⁴⁴

Turning to the dimension of “strength,” Tushnet was keen to stress that strong- and weak-form systems do not necessarily differ either in terms of the scope of review ⁴⁵ or the “degree to which courts are aggressive in enforcing their views of fundamental rights or in whether the courts defer or refuse to defer to legislative judgments about whether legislation is consistent with constitutional guarantees.” ⁴⁶ He acknowledges that both the scope and intrusiveness of review could be the same in strong-form and weak-form systems. ⁴⁷ It follows that, for Tushnet, the dimension of strength follows from the dimension of form . Strong-form review is strong because judicial review is structured so that courts can displace the legislature using powers which are coercive, ⁴⁸ conclusive, ⁴⁹ and unrevisable by ordinary legislative majorities. ⁵⁰ In contrast, weak-form review is weak because court decisions are explicitly revisable in this way—they can be displaced by an ordinary legislative majority. Thus, the strength or weakness of both systems is cashed out largely in terms of which institution can ultimately displace the other as a matter of formal constitutional design. It does not depend on how interventionist or passive courts are in the exercise of their reviewing role either in particular cases or with respect to specific issues.

As we shall see later, there is some instability in this claim and, indeed, some suggestions from Tushnet himself that constitutional traditions of deference (either by the legislature to the courts or the courts to the legislatures) will have a bearing on the characterization of the system as “weak-” or “strong-form.” ⁵¹ Nonetheless, for present purposes, it suffices to note that, when stated at its highest level of abstraction, Tushnet intends the distinction to mark out a difference in constitutional design, not in the intensity with which the courts’ reviewing powers are exercised. Let us now turn to an examination of how this distinction looks when it is applied to the UK Human Rights Act 1998. This is an important inquiry, because the UK system is often portrayed (both by Tushnet and others) as a leading example of “weak-form review.”

3. The Human Rights Act 1998: form and function

When the Human Rights Act 1998 was being enacted, the key concern across the political spectrum was to find a way of allowing judges to enforce human rights standards while preserving the notion of parliamentary sovereignty. ⁵² There was no political appetite for giving the courts a strike-down power on the American model. ⁵³ The challenge was to find a way of protecting rights, while ensuring that Parliament would ultimately have the “final say” in determining what the law should be. ⁵⁴ How could this be achieved? The drafters came up with a novel “compromise solution” ⁵⁵ which they embedded into the HRA’s structure and design. ⁵⁶

The HRA contains three mechanisms designed to enhance the protection of human rights in primary legislation. The first two are powers given to the courts; the third is a mechanism oriented towards parliamentary procedure. The powers given to the courts are interlinked. The first is an interpretive duty contained in § 3 HRA which reads: “so far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights.” ⁵⁷ However, if a court decides that a rights-consistent interpretation is not possible, this triggers the second judicial mechanism for protecting rights, namely, the power to issue a “declaration of incompatibility” under § 4 HRA. Crucially, this declaration “does not affect the validity, continuing operation or enforcement of the provision in respect of which it is given,” ⁵⁸ and is not binding on the parties to the proceedings in which it is made. ⁵⁹ It has no binding legal effect.

Turning to the impact of the HRA on parliamentary procedure, § 19 HRA provides that, when introducing legislation into Parliament, ⁶⁰ a Minister must make (a) a statement to the effect that in his view the provisions of the Bill are compatible with the Convention rights (“a statement of compatibility”); or (b) a statement to the effect that although he or she is unable to make a statement of compatibility the government nevertheless wishes the House to proceed with the Bill. Section 19 specifically allows the Government to proceed with legislation even when it believes that it may violate Convention rights, but it obliges the Government to make such an intention explicit during the passage of the Bill through Parliament. ⁶¹ Although there is no legal obligation on a proposing Minister to give reasons for the § 19 statement or even to initiate parliamentary debate on the issue, ⁶² a practice has now developed whereby the government provides a Memorandum outlining its reasons for its views on compatibility. ⁶³ Ministers almost always make a positive statement of compatibility. A negative “statement of in compatibility” has only been made on one single occasion (in relation to the Communications Bill 2003), and that statute was subsequently upheld by the UK Supreme Court as compatible with Convention rights. ⁶⁴

From this brief outline, it is easy to identify some key features which are salient for the distinction between strong-form and weak-form review. Most notably, there was a deliberate decision not to give British judges the power to strike down legislation, and Parliament is expressly allowed to enact legislation which seems to contravene Convention rights. Therefore, as a matter of constitutional design, the HRA seems to fit squarely within Tushnet’s typology as an instance of “weak-form review.” Let us now look at how these provisions operate in practice, in order to see whether this is a useful characterization. Of crucial significance will be the question whether these mechanisms succeed in fulfilling what Tushnet takes to be the underlying aim of “weak-form review,” namely, to give the legislature “the last word” on issues concerning human rights. ⁶⁵ I will focus primarily on the two judicial powers and the political response to them, because these are the most salient features with respect to Tushnet’s distinction between strong-form and weak-form review. Therefore, I will not discuss the various practices of “legislative rights review” which are undertaken in Parliament during the legislative process ⁶⁶ or comment on the important work carried out by the parliamentary Joint Committee on Human Rights in raising awareness about rights within Parliament. ⁶⁷

3.1. Interpretation under § 3 HRA

The crucial question surrounding § 3 HRA is how far can judges go when interpreting legislation compatibly with Convention rights. ⁶⁸ The terms of § 3 do not provide much guidance, stating simply that courts should adopt a Convention-compatible reading “so far as it is possible to do so.” The courts have settled on an expansive understanding of § 3 which allows them to alter the interpretation of otherwise unambiguous statutory provisions. They have done so by adopting a two-stage approach to the interpretive obligation, which functions in the following way. ⁶⁹ The first step is to establish whether the legislation, as ordinarily understood, violates Convention rights. If it does not, that is the end of the matter. However, if the courts believe that there is a prima facie infringement of Convention rights, then the courts go on to see whether it is nonetheless possible to find a rights-consistent interpretation under § 3(1) HRA which could remove or “cure” the apparent incompatibility. ⁷⁰ At this second stage, the courts have held that § 3 allows them “to adopt an interpretation which linguis tically may appear strained. The techniques to be used will include the reading down of express language in a statute but also the implication of provisions.” ⁷¹

The leading case on interpretation under § 3(1) is Ghaidan v. Mendoza.⁷² This case concerned the rights of a surviving same-sex partner of a tenant to a statutory tenancy under the Rent Act 1977. The Rent Act granted a right of succession to the “surviving spouse” of a tenant, where “spouse” was defined as “a person who was living with the original tenant as his or her wife or husband.” ⁷³ The question was whether this provision applied to same-sex partners. The House of Lords held that, as ordinarily understood, the Rent Act did not apply to same-sex couples and therefore clearly violated Mr. Mendoza’s Convention rights. ⁷⁴ The next question was whether this apparent or prima facie infringement could be eliminated by adopting a § 3(1) interpretation. By a majority of 4:1, the House of Lords held that it could. How was this achieved? The House of Lords held that while the language of the Rent Act seemed to preclude same-sex couples, the “social policy” underlying the Act was to give secure tenancies to couples who were in a stable and loving relationship and this policy could be applied equally to same-sex couples. In this way, the courts used § 3 to read the Rent Act compatibly with Convention rights, thus removing its potentially discriminatory effect on same-sex couples. ⁷⁵

All judges agreed that the interpretive obligation in § 3(1) was “of an unusual and far-reaching character” ⁷⁶ and was not confined to resolving ambiguities in the legislation being interpreted in favor of a Convention-compatible meaning. ⁷⁷ Giving the leading judgment, Lord Nicholls clarified that

even if, construed according to the ordinary principles of interpretation, the meaning of the legislation admits of no doubt, section 3 may nonetheless require the legislation to be given a different meaning . . . Section 3 may require the court to depart from the legislative intention, that is, depart from the intention of the Parliament which enacted the legislation. . . . ⁷⁸

What interpretive tools could the courts use to make legislation compliant with Convention rights? Lord Nicholls offered some guidance:

Section 3 enables language to be interpreted restrictively or expansively. But section 3 goes further than this. It is also apt to require a court to read in words which change the meaning of the enacted legislation, so as to make it Convention-compliant. In other words, the intention of Parliament in enacting section 3 was that, to an extent bounded only by what is “possible,” a court can modify the meaning, and hence the effect, of primary and secondary legislation. ⁷⁹

Although this expansive approach to the interpretive obligation is not without its academic critics, ⁸⁰ it has been fully endorsed in subsequent cases and is now well settled in the jurisprudence. As Lord Bingham stated quite simply in R. (Al-Skeini) v. Secretary of State for Defence : “section 3 provides an important tool to be used where it is necessary and possible to modify domestic legislation to avoid incompatibility with the Convention rights protected by the Act.” ⁸¹ There is now “no room for doubt” ⁸² that the interpretative obligation under § 3 is “a very strong and far-reaching one, and may require the court to depart from the legislative intention of Parliament.” ⁸³

Using § 3, the courts have adopted bold and creative interpretations. There have been a number of cases where the courts have used § 3 to “read down” provisions which clearly imposed a legal burden of proof on the defendant, to mean that the defendant was only obliged to discharge an evidential burden of proof. ⁸⁴ In R. (Hammond) v. Home Secretary , ⁸⁵ the House of Lords treated a statutory provision that a life prisoners tariff “is to be determined by a single judge of the High Court without an oral hearing ” as giving the judge a discretion to require an oral hearing where fairness required it in order to render it compatible with the right to a fair trial. ⁸⁶ In the most controversial case decided in the early days of the HRA ( R v. A ), ⁸⁷ the House of Lords held that a rape-shield provision enacted by Parliament just the previous year, ⁸⁸ which severely curbed judicial discretion to admit sexual history evidence of the complainant in rape trials (including sexual history evidence with the accused), violated the accused’s right to a fair trial. Although the relevant statutory provision was extremely detailed and clearly excluded the admissibility of such evidence (indeed was enacted specifically in order to curb judicial discretion to admit such evidence), ⁸⁹ the court read an entire subsection into the Act in order to give judges the discretion to decide whether such evidence was “so relevant to the issue of consent that to exclude it would endanger the fairness of the trial under Article 6 of the Convention.” ⁹⁰ Though “linguistically strained,” this implication of provisions was deemed permissible under § 3(1), largely in order to ensure that the accused received a fair trial. ⁹¹

A number of preliminary points can be made here about the nature of the interpretive obligation under § 3 HRA. The first is that whatever else we may say about it, it seems inapt to describe it as weak. As interpreted by the higher courts in the UK, it has been understood as a “strong adjuration” ⁹² to render legislation compatible with Convention rights even if it goes against the unambiguous wording and clear intention of the statute. The dynamics of the two-stage approach makes plain that § 3 is only engaged if there is a prima facie violation of Convention rights. ⁹³ The question at the second stage then becomes whether the statute can nonetheless (i.e., despite initial appearances to the contrary) be “ rendered Convention-compatible” ⁹⁴ using the tools of interpretation. The courts ask themselves what they can do to “ cure the incompatibility” ⁹⁵ which would otherwise exist, either by deleting or inserting words into the statute. ⁹⁶ Indeed, it is precisely because of its strength that the judicial use of § 3 has been accused of being indistinguishable from legislative amendment. ⁹⁷

Second, recall that in mapping out his distinction between strong-form and weak-form review, Tushnet characterized rights-consistent interpretation as “the weakest variant of weak-form review.” ⁹⁸ Weak-form review allows courts to “interpret but not displace legislation.” ⁹⁹ But this underestimates the extent to which various modes of statutory interpretation can themselves “displace” or at least modify clear legislative provisions. Transformative interpretations under § 3 HRA certainly seem to displace the clear language of unambiguous statutes and replace them with a rights-consistent interpretation of the court’s own making. In fact, it is arguable that, at least in some cases, § 3 allows the UK courts to engage in a stronger form of displacement than that which normally occurs when courts invalidate legislation in those jurisdictions with a “strike-down” power. After all, “striking down” is a metaphor for the power of courts to declare that legislation violates rights, which has the effect of rendering the legislative provision void or inapplicable. ¹⁰⁰ When this occurs, it is then up to the legislature to decide how to amend the legislation in order to rectify the rights-violation or whether to abandon the legislation altogether. ¹⁰¹ However, when a UK court adopts a § 3 interpretation, it does not confine itself to the negative task of merely identifying the existence of a rights-violation. It also takes the further (positive) step of implementing corrective action to rectify that rights-violation itself, using various interpretive techniques like reading in, reading down and “the implication of provisions.”

Furthermore, it is hardly clear that if judges effectively re-write statutes or rectify them to some degree in order to “cure” them or save them from interfering with rights, that this is less worrying from a democratic point of view than if judges invalidate the legislation. ¹⁰² At least judicial invalidation gives the legislature the opportunity to decide what to do next to remedy the rights-violation, ¹⁰³ thus inviting more collab oration and input from the other branches than a rectifying interpretation. Indeed, in many jurisdictions with a strike-down power, judges sometimes suspend the effects of the strike-down precisely to give the legislature more time to adapt to the ruling and work out the most appropriate remedial solution. ¹⁰⁴ This expands the room for maneuver available to the legislature when deciding how to craft corrective legislation.

But if judges rectify the statute themselves by way of judicial interpretation under § 3, they effectively engage in a form of reconstructive surgery on the legislation which cures the rights-violation with immediate effect. ¹⁰⁵ It is a form of judicial amendment which involves some displacement of the will of Parliament. ¹⁰⁶ And in most cases, this displacement is effectively final because Parliament bears a burden of legislative inertia if it wishes to override a § 3 interpretation. ¹⁰⁷ In fact, the Westminster Parliament has never overcome that inertia threshold to enact legislation to overturn or modify a § 3 interpretation issued by the higher courts. ¹⁰⁸ Thus, if the defining feature of weak-form review is the relative ease with which the legislature can respond to judicial decisions on rights, this has not been borne out in practice in the UK, at least as far as § 3 interpretations are concerned.

3.2. From interpretation to declaration

Let us now turn to the declaration of incompatibility which is often thought to be the most innovative design-feature of the UK HRA. The first point to note about the de claration of incompatibility is that it is inextricably linked with the interpretive mandate under § 3. This is because the courts can only issue a declaration “if construction cannot resolve the incompatibility.” ¹⁰⁹ Therefore, the limits of § 3 trigger the declaration under § 4. The HRA does not specify what those limits are. Therefore, it is up to the courts to articulate the grounds on which they would decide that a rights-consistent interpretation is not possible.

In Ghaidan , the House of Lords outlined two main limits. The first is that “the courts should adopt a meaning inconsistent with a fundamental feature of the legislation.” The second is that the courts should not make decisions “for which they are not equipped. There may be several ways of making a provision Convention-compliant, and the choice may involve issues calling the legislative deliberation”. ¹¹⁰ We can call these “the fundamental features limit” and the “legislative deliberation limit” respect ively. ¹¹¹ They indicate that even when it is possible for the courts to read down or read in words to statutes using § 3(1), they must not go against “the entire substance” ¹¹² of the legislation being construed. ¹¹³ Where a rights-consistent interpretation would cause too much discordance within the legislative scheme or require comprehensive legal reform, then the courts will tend to issue a declaration of incompatibility, leaving it to Parliament to remedy the rights-violation. ¹¹⁴

In general, the courts have “declined to try to formulate precise rules” ¹¹⁵ on when the limits of § 3 have been reached, but they have expressed confidence that, in practice, such cases will be “fairly easy to identify.” ¹¹⁶Bellinger v. Bellinger¹¹⁷ provides a good illustration of the grounds on which the UK courts may decide to issue a declaration of incompatibility, rather than rectify the rights-violation themselves using § 3. Mrs. Bellinger was a post-operative male to female transsexual who argued that the Matrimonial Causes Act 1973, which only allowed a “male” and “female” to marry (with gender determined at birth) violated her right to a family life and was discriminatory. The House of Lords agreed that, as ordinarily understood, the 1973 Act violated Mrs. Bellinger’s Convention rights. Although they could have easily read in words to include transsexuals in the 1973 Act, ¹¹⁸ thus remedying the rights-violation with immediate effect, they declined to do so. Opting for a declaration of incompatibility, they explained that giving “male” and “female” an extended meaning

. . . raises issues whose solution calls for extensive inquiry and the widest public consultation and discussion. Questions of social policy and administrative feasibility arise at several points, and their interaction has to be evaluated and balanced. The issues are altogether ill-suited for determination by courts and court procedures. They are pre-eminently a matter for Parliament. ¹¹⁹

This was an issue “calling for comprehensive legislative reform and not piecemeal judicial development.” ¹²⁰ Therefore, a declaration of incompatibility was the most appropriate solution.

A number of points can be made about this interplay between § 3 and § 4. The first is that the choice between judicial rectification of an incompatible law under § 3 and the declaration of incompatibility mechanism under § 4 is context-dependent, heavily influenced by remedial concerns. ¹²¹ If the courts believe that they can remedy the rights-violation by way of interpretation of § 3 (thus doing justice to the individual litigant) without causing undue discordance in the legislative scheme under scrutiny or causing any other negative side-effects, then they will opt for a § 3 interpretation. ¹²² However, if root-and-branch reform of the whole area is required and the courts believe that they have insufficient institutional competence and legitimacy to rectify the problem, then they will tend to opt for a § 4 declaration. ¹²³ As Lord Bingham observed extra-judicially, in choosing between § 3 or § 4, the courts should assess

whether, faced with the incompatibility, Ministers and Parliament would have practically- effective options of how to cure it: if so, they should be given the opportunity to find the best remedy; if not, s.3 provides the neatest and most final conclusion. ¹²⁴

Second, the UK courts have enunciated a general principle that § 3 will be treated as the primary remedial measure under the Act, with declarations of incompatibility only used as a last resort. ¹²⁵ This is now a settled principle in the case law. ¹²⁶ As the Supreme Court has held on numerous occasions: “section 3(1) is the primary remedy and resort to making a declaration of incompatibility must always be an exceptional course .” ¹²⁷ Why have the courts given primacy to § 3 in this way? One reason might be the fact that when the Human Rights Bill was being debated in Parliament, Government Ministers expressed their beliefs that judges would be able to interpret legislation compatibly with the Convention “in almost all cases.” ¹²⁸ But the main reason is that only a § 3 interpretation can give the individual litigant a remedy. So, ironically, it is precisely the remedial emptiness of the declaration of incompatibility which may lead the UK courts to adopt highly creative and interventionist interpretations in order to fill the remedial gap. In some of the most controversial cases of judicial rectification under § 3, the courts were clearly motivated by a concern that prospective legislation following a declaration of incompatibility would leave individual litigants empty-handed and cause them substantial injustice. Thus, in a case concerning the right to a fair trial for the charge of rape, the courts were prepared to engage in a substantial rewriting of the legislative provision at issue precisely because the accused would otherwise be at risk of receiving an unfair trial. ¹²⁹ In other words, some statutory distortion was the price to be paid for remedying the potential injustice in this individual case. So the fact that the declaration of incompatibility has no immediate legal effect, can create strong incentives for British courts to adopt bolder more creative statutory interpretations than would generally be the case in systems where the courts have the power to invalidate legislation. ¹³⁰

There is one more factor which bears mention here and goes directly to the distinction between strong- and weak-form review. Recall that as Tushnet and other scholars present it, the distinctively attractive feature of weak-form review (and its main advantage over strong-form review) is that it gives the legislature the “last word” on questions concerning rights. Experience with the UK HRA shows that, quite often, the government does not want the last word and is quite happy to let the courts make the decisions on these questions, thus obviating the need for the government to rectify the problem. ¹³¹ Thus, when defending challenges to primary legislation under the HRA, lawyers for the UK Government often argue that the court should rectify any rights-violation by way of § 3, rather than relying on a declaration of incompatibility. ¹³² Commenting on this practice extra-judicially, Lord Phillips observed that in his judicial experience, the bold and creative use of § 3 has

suited [Government Ministers] rather well. Ministers do not like declarations of incompatibility. Provided that the main thrust of their legislation is not impaired they have been happy that the courts should revise it to make it Convention-compliant, rather than declare it incompatible. In my experience, Counsel for the Secretary of State usually invites the court to read down, however difficult it may be to do so, rather than make a declaration of incompatibility. ¹³³

Why do Governments often argue for a § 3 interpretation and let the courts do their human rights handiwork for them? Why do they not rush to seize the “last word” on such important questions, as many constitutional theorists believe they should? One possibility is that the Government does not want to incur the political costs involved in the headline-grabbing damning verdict involved in a declaration of incompatibility, and would prefer the courts to modify the law through the relatively invisible means of judicial interpretation. ¹³⁴ After all, to the general British public with no detailed understanding of the nuances of the HRA, a § 3 interpretation seems like a clean bill of health in human rights terms. This is natural since the practical effect of a case which relies on § 3 is that the legislation no longer violates rights (because the potential rights-violation is removed or “cured” by way of interpretation). ¹³⁵ On the other hand, it is harder to present a declaration of incompatibility as anything other than a straightforward failure by the Government to protect human rights. ¹³⁶ Therefore, although the appeal of weak-form systems is thought to be that they give the legislature “the last word” on rights, the fact is that the legislature may not want the last word, because it does not want to bear the brunt of political criticism that it does not respect rights. The UK experience shows that governments may prefer to leave controversial decisions about rights to the courts where they can be dealt with under the political radar, rather than engaging in an open political dialogue on these questions for all to see.

3.3. The effect of the declaration of incompatibility

Let us now turn to examine the nature and impact of the declaration of incompatibility. As we saw earlier, a key feature of the declaration of incompatibility is that it is not binding on the parties and offers litigants no remedy. That is certainly a weakness of the declaratory mechanism and a reason why most litigants will prefer a § 3 interpretation. But is the declaration of incompatibility weak overall? This is a crucial question, since much of the animus of the strong-form/weak-form distinction hinges on the fact that in strong-form systems of review, the courts can strike down legislation in a decision which is final and coercive, whereas under weak-form review, the courts are only give so-called “weaker remedies” like the declaration of incompatibility, thus leaving the last word with Parliament.

Although the declaration of incompatibility is not legally binding on the Government or legislature, it is not as weak as may at first appear for a number of reasons. First, there is the crucial international law dimension in the scheme of the HRA, which must not be overlooked. ¹³⁷ If a British court issues a declaration of incompatibility which the government and/or Parliament fail to remedy, the aggrieved litigant is likely to bring an application to Strasbourg. There, the applicant will argue that his or her Convention rights have been violated (as confirmed by the highest domestic courts) and that the UK government has nonetheless failed to remedy the violation. In this situation, it is highly likely that the Strasbourg court will find a rights violation, ¹³⁸ whereupon the UK Government is placed under an international law obligation to change the law in light of this judgment. ¹³⁹

Therefore, the Government knows that if it defies a declaration of incompatibility, it will only succeed in postponing the problem, since it is almost inevitable that it will lose its case in Strasbourg, with all the adverse publicity (both domestic and international) that this would entail. The risk of suffering significant political embarrassment if it loses its case in Strasbourg creates a strong incentive to comply with a declaration—all the more so, since it was partly to avoid such embarrassment that the British Parliament adopted the HRA in the first place. ¹⁴⁰ Therefore, the UK Government has a lot to lose and potentially very little to gain by defying a declaration. Although the Declaration of Incompatibility (DOI) is often portrayed as a purely political mechan ism, it also has significant legal repercussions. Although it is not, in itself, legally binding, it is a means by which the courts can alert Parliament to a binding obligation in international law. ¹⁴¹

Second, there are multiple sources of political pressure on the UK Government to comply with declarations of incompatibility. Not only is there the problem of adverse publicity attracted by a judicial ruling declaring that legislation violates rights, these rulings are often seized upon by Opposition MPs to galvanize opposition to the Government’s policy within Parliament. ¹⁴² Furthermore, declarations are often relied on by the parliamentary Joint Committee on Human Rights to raise the profile of the human rights violation and to strengthen the political pressure on the Government to remedy the rights-violation, both within and outside Parliament. ¹⁴³ All told, the political repercussions of resisting a judicial finding of a rights violation, combined with the legal repercussions in the (highly likely) event of an adverse finding from Strasbourg, set against the backdrop of the traditional comity between Parliament and the courts and the general respect for court decisions, means that a declaration of incompatibility can have a much stronger practical force than its legally non-binding status might suggest. ¹⁴⁴

Third, when the Human Rights Bill was being introduced in Parliament, many members of the Government stated in parliamentary debate that there would be a strong expectation that Government would comply with the declarations. Thus, the then Lord Chancellor stated that “we expect that the government and Parliament will in all cases almost certainly be prompted to change the law following a declaration of incompatibility.” ¹⁴⁵ This expectation, described by some commentators as a “constitutional expectation,” ¹⁴⁶ that the Government/Parliament will remedy the rights-violation following a declaration of incompatibility, ¹⁴⁷ has been generally borne out in practice.

Since the enactment of the HRA, there have been 21 final declarations of incompatibility and in almost every single case, the Government/Parliament has responded in some way to remedy the rights-violation. ¹⁴⁸ It is true that compliance has sometimes been minimal and begrudging, and in recent times has been accompanied by excoriating political rhetoric about the problems of court rulings (particularly those of the Strasbourg court) on Convention rights. ¹⁴⁹ Moreover, there has been considerable delay in implementing declarations. ¹⁵⁰ But there has been “no case to date where the Government or Parliament affirmatively chose not to remedy incompatible legislation.” ¹⁵¹ Despite delays and some negative political rhetoric, there is a near-perfect rate of compliance with declarations of incompatibility since the HRA was enacted. ¹⁵²

The only possible exception concerns a declaration of incompatibility concerning a ban on prisoner voting where the Government has delayed almost ten years in implementing any legislative change on this issue, despite a number of rulings from the Strasbourg court affirming that a blanket ban on prisoner voting violates the Convention. ¹⁵³ The issue of prisoner voting is marked by special features. The first is that there is cross-party support for denying prisoners the right to vote across all the main political parties in the UK. The second is that it is believed that the opposition to prisoners getting the right to vote incurs no electoral cost, since the view of the political elites is said to chime with popular opinion. Both of these conditions are rarely achieved. Therefore, if this is the threshold at which a declaration of incompatibility is defied or ignored, then it is an extremely high one indeed. It will only be met in exceptional circumstances. For sure, the political rhetoric on this issue is sometimes defiant, with politicians suggesting that they will ignore or defy the Strasbourg ruling or even withdraw from the Convention altogether. Nonetheless, a parliamentary Select Committee has recommended that the Government should implement the ruling in a minimal way. ¹⁵⁴ It remains to be seen how the Government will respond. ¹⁵⁵ As things stand, it continues to do nothing.

The high rate of compliance with declarations of incompatibility has led some commentators to suggest that a constitutional convention may be developing whereby Parliament perceives itself to be under a constitutional obligation to comply with DOIs. ¹⁵⁶ For a constitutional convention to exist, that there must be a common political practice which is treated by the key institutional actors as politically (though not legally) binding. ¹⁵⁷ While it may be too early to come to a definitive conclusion on whether compliance with declarations of incompatibility has now reached the status of a constitutional convention, there is growing evidence to support it. Certainly, there is “an almost uniform practice of legislative amendment by way of response [to de clarations].” ¹⁵⁸ Moreover, many leading politicians and journalists in the UK describe the decisions relying on the declaration of incompatibility as binding or as having the effect of “overturning” or “striking down” primary legislation. ¹⁵⁹ Even when MPs express considerable frustration with a declaration of incompatibility, they present it as “an authoritative finding, not merely as opinion or advice,” ¹⁶⁰ suggesting that they are bound to comply with it. ¹⁶¹

One explanation for this is advanced by Jan Van Zyl Smit. He argues that it may be a crude form of political buck-passing which allows Ministers to avoid the blame for unpopular decisions while directing public criticism towards the courts. ¹⁶² A less cynical explanation is also possible. It may simply be that in countries where the courts are held in high esteem and there is a relatively strong belief in the import ance of protecting rights and the rule of law within the constitutional culture, all judgments by the highest courts (whether declaratory or not) will effectively be treated as binding or the very least highly persuasive and worthy of respect in most cases. ¹⁶³ If this is true, then the novel design mechanisms associated most prominently with the model of “weak-form review” (namely the UK’s declaration of incompatibility or Canada’s notwithstanding clause) are unlikely to fulfill their promise of giving the legislature “the last word” on questions concerning rights—at least in the sense of enabling elected representatives to disagree freely, regularly and overtly with a high-profile court ruling concerning rights. This explanation also tends to support the view that a constitutional convention of compliance is developing.

It may also cast doubt on the plausibility of portraying systems like the UK as embodying a “dialogue model of rights-protection,” ¹⁶⁴ at least if the metaphor of dialogue is taken to imply an open conversation among equals combined with the fallback position of legislative finality. ¹⁶⁵ The empirical realities do not match this metaphorical picture. In general, the UK Parliament accepts court-rulings about whether Convention rights have been violated in particular cases, while seeing its own role as fashioning the appropriate remedy for the rights-violation in ways which are not precluded by the judicial decision. ¹⁶⁶ This is less like an open conversation and more like a complex division of labor where each branch of government performs different (though complementary) roles in protecting rights.

The strength of the various pressures on Parliament to amend legislation to comply with declarations of incompatibility has led some academic commentators on the Act to suggest that although the UK courts are not empowered to strike down legislation, they are “effectively doing so indirectly, given that government would almost certainly want to change the law.” ¹⁶⁷ Though in theory hortatory, in practice a declaration of incompatibility resembles “a de facto judicial power to procure the amendment of legislation which unlawfully qualifies fundamental rights” akin to a strike-down power. ¹⁶⁸ Lord Hoffmann has suggested extra-curially that the difference between a judicial strike-down power and the declaration of incompatibility is merely “technical,” because “if the courts make a declaration of incompatibility, the political pressure upon the government and Parliament to bring the law into line will be hard to resist.” ¹⁶⁹ Certainly, if the declarations of incompatibility are complied with in almost every case, then they have a similar effect as a strike-down power in terms of ultimately securing prospective legislative change in line with judicial pronouncements on what rights require. ¹⁷⁰

The main difference between the powers lies in the fact that the declaration of incompatibility has no immediate impact on the legislation and gives no immediate benefit to the litigants. As we have seen, the lack of retroactive relief following a declaration of incompatibility is not without significance. It can incentivize the courts to adopt a rights-consistent interpretation—even when that strains the language of the statute. ¹⁷¹ It may also disincentivize potential litigants from bringing human rights challenges at all in those cases where the rights-violation is deeply embedded in a fundamental feature of the legislative framework, thus precluding a § 3 interpretation. But the overall effect is to strengthen rights-protection through interpretive creativity, rather than leave a lacuna where rights are not protected by the courts at all. It changes the form of judicial rights-protection, but does not necessarily make it weaker. In fact, an argument can be made that—contrary to Tushnet’s taxonomical assumptions—it gives judges a stronger hand in protecting rights than they have in strong-form systems.

4. Is the UK HRA an example of weak-form review?

There is no doubt that when the Human Rights Act was being drafted, a conscious effort was made to ensure that judges would not have the power to strike down primary legislation and that Parliament would be free to take its own view of what rights required and even to enact legislation which seemed to violate Convention rights, if it so wished. ¹⁷² Therefore, if the distinctive feature of weak-form review is that, under such a system, court rulings “cannot displace the ultimate power of legislatures to determine public policy,” ¹⁷³ it seems plausible to suggest that, at least at the level of formal institutional design, the UK system is an instance of weak-form review. ¹⁷⁴

However, when we drill down into the detail of how the HRA works in practice, a different picture emerges. For one thing, there is the enormous strength of the interpretive duty as a way of giving (immediate) legal effect to Convention rights in primary legislation. Using § 3, judges “cure” rights-violations by rectifying statutory language. As we have seen, declarations of incompatibility are used as a last resort in cases concerning rights. They have been complied with in almost every case. Furthermore, when we turn to examine the practice surrounding § 19 HRA, it transpires that although the Government is entitled to introduce legislation which it believes to contravene Convention rights, it has only done so in one instance. In general, all legislative proposals introduced by the Government are accompanied by a “Statement of Compatibility,” rather than a negative “Statement of In compatibility.” ¹⁷⁵

So, when we return to Tushnet’s typology of strong-form and weak-form review, the UK system seems to pose something of a classificatory challenge. It is true that, as a matter of constitutional design, it seems to be structured so as to give the legislature the possibility of having “the last word.” But, in practice, court rulings have been treated as authoritative and binding on the other branches and have tended to prevail. The legislature has not been keen to rely on the mechanisms available to it under the HRA to defy or overrule or ignore court decisions. If the distinctive feature of strong-form review is that the decisions of the courts have “normative finality” on questions of rights, ¹⁷⁶ then the UK has a strong claim to be an instance of strong- rather than weak-form review.

Of course, a sophisticated comparative law scholar like Mark Tushnet is not oblivious to the developing practice in the UK. Writing just as the HRA was enacted, Tushnet argued that “the HRA system would be indistinguishable from strong-form review if statutes that courts declared incompatible with the European Convention were always amended to remove the incompatibility.” ¹⁷⁷ Tushnet was fully apprised of the import ant international law dimension of the Declaration of Incompatibility, observing that “the structure of European human rights law might make the British courts’ predictive judgments self-fulfilling, in which case their declarations of incompatibility would be the equivalent of exercises of strong-form review.” ¹⁷⁸ He acknowledged that “the political dimension of even the interpretive mandate, the weakest variant of weak-form judicial review, suggests that the difference between weak-form and strong-form review may not be as dramatic as it might seem at first.” ¹⁷⁹ In short, he realized that a distinction along the dimension of form, might not carry through to a meaningful distinction along the dimension of strength.

In fact, Tushnet was one of the first scholars to point out that “weak-form review” may be vulnerable to potential (though not inevitable) instabilities, such that it might escalate into strong-form review in some contexts or, alternatively, degenerate into a pure legislative supremacy model. ¹⁸⁰ He argued that the distinction would simply collapse in situations where legislatures failed to challenge or override court rulings: “if legislatures routinely accede to court decisions, weak-form review is simply strong-form review in disguise.” ¹⁸¹ This led him to set the following criterion of success for weak-form systems, namely, that

Legislatures in successful weak-form systems must sometimes, and not rarely, respond to judicial interpretations by asserting their own, contrary understandings of the constitution . . . the reason is that the system is effectively strong-form in the absence of legislative responses. ¹⁸²

Given that the UK Parliament rarely (if ever) responds in a way which rejects court rulings outright, we may view the UK system as “effectively strong-form” within a nominally “weak-form system.” Or we may think of it as an unsuccessful instance of “weak-form review.” At times, Tushnet goes so far as to suggest that it is simply indistinguishable from strong-form review.

The question which then arises is whether it is useful or illuminating to characterize the UK system in this way. I believe that it is problematic and misleading to conceptualize the UK system as a failed instance of weak-form review, or as one which is “weak in form, but strong in practice.” In the next section, I will argue that the difficulty of classifying the UK system is rooted in deeper problems and instabilities inherent in the distinction itself—problems which cast doubt on the usefulness of the distinction as a typological tool for comparative constitutional law.

5. Collapsing the distinction: from polarized to blended systems

The first problem with the distinction between strong-form and weak-form review is that it invites us to characterize systems using the metric of strength in a way which is largely based on its formal institutional design, divorced from the way in which that design is implemented in practice. Thus, Tushnet is keen to emphasize that the strong-form/weak-form distinction is one between different ways of structuring or institutionalizing powers of judicial review, rather than on the basis of how deferential one institution is to the other. In one sense, it may be harmless to characterize and distinguish systems of review based on the formal powers given to the courts and legislature. However, the limitations of this kind of exercise should also be immediately apparent. As we all know, function does not necessarily follow form in constitutional law. There are many systems where courts have the power to strike down primary legislation, but they never use their powers or, if they do, they exercise them in a timid and supine fashion. It seems artificial and misleading to characterize such systems of judicial review as “strong” or “strong-form,” even if the actual power of the courts in these systems is weak, if not non-existent.

Tushnet is aware of this artificiality and it leads him to incorporate the actual real-life dynamics of the legislature-court relationship into his characterization of various systems. Thus, he describes systems like Japan (where the courts possess the power to strike down legislation, but hardly ever use it) as “quasi-weak-form-review within formally strong-form systems.” ¹⁸³ And he describes the Indian system as one of “quasi-weak-form-review” rather than strong-form review (despite the fact that the Indian Supreme Court is known as one of the activist courts in the world), because the Indian Parliament often relies on constitutional amendments to respond to judgments of the Indian Supreme Court. ¹⁸⁴ When faced with the task of classifying such systems (which he characterizes as formally strong but weak in practice), Tushnet is reluctant to eschew considerations of how the various institutional actors actually use their powers. Put otherwise, he shows a reluctance to peg the metric of strength or weakness closely to the formal design, because he can see that this way of measuring the strength of a system can produce counter-intuitive results and risks distorting or mischaracterizing many systems. Quite rightly, he joins Rosalind Dixon in conceding that

substance as well as form matters when it comes to judge the strength of judicial review. To say that review is strong does not mean that constitutional review is strong in practice, taken to mean that the constitutional court regularly sets its constitutional interpretations against those implicit in the legislation it reviews. ¹⁸⁵

But once we introduce the strength or weakness of the powers as they are exercised in practice, the distinction begins to lose some of its purchase as a useful typological tool for identifying different systems of rights-based review in comparative constitutional law. The dimension of “form” begins to drop out and be subsumed by a more rounded evaluation of “strength” based on an overall assessment of how the system works in practice. The real differences in the strength of the systems may result from factors which have less to do with the formal design and more to do with the actual powers of the various institutional actors within the system and the broader political, legal and strategic dynamics of the interaction between them. In short, the real differences may be ones of constitutional culture not constitutional form. ¹⁸⁶

This leads on to the second (related) problem with the distinction between strong-form and weak-form review. This is that the distinction pivots crucially on the idea of institutional “displacement.” In a system of strong-form review, Tushnet says that the courts “displace” the legislature, whereas in weak-form review, the legislature can displace the courts. ¹⁸⁷ One problem with relying so heavily on the metaphor of displacement (especially when it is combined with an archetypal reliance on the US as the standard-bearer of strong-form review) is that it encourages us to characterize a system as weak or strong largely on the basis of whether it includes a judicial strike-down power. In other words, the distinction pegs the metric of strength closely to the power of judicial invalidation of legislation on the American model. The risk with using this as a typological tool and transposing it to other systems is that it may obscure from view other significant (albeit less overt) ways in which the strength of the judicial power can be measured and felt.

As can be seen vividly using the UK example, doctrines of interpretation can be a strong and effective way of ensuring that legislation complies with rights. Though less eye-catching and dramatic than a strike-down power, these powers are often no less potent as a tool of rights-protection. At the very least, it seems problematic to pre-judge an interpretive mandate like § 3 as “the weakest variant of weak-form review” ¹⁸⁸ when it has such a powerful ability to remove and rectify rights-violations. Not only should we guard against the fallacy that interpretive tools are weak, we should also be alert to the possibility that, ironically, they may give the courts more control over legislation than a negative power of invalidation. ¹⁸⁹

Moreover, if we look at the UK system through the lens of the distinction between strong-form and weak-form review, we are drawn to focus primarily on the contrast between the (coercive) strike-down power and the (non-binding) declaration of incompatibility. ¹⁹⁰ If that is the central or pivotal contrast, then it is indeed tempting to conclude that the US is strong and the UK system is weak. However, the UK example shows that this focus is too narrow and potentially misleading. The remedial centerpiece of the UK system is actually § 3 (not § 4) and the declaration of incompatibility is used as a measure of last resort. Therefore, rather than being the central or defining feature of the UK system of rights-protection, the declaration of incompatibility is in fact a subsidiary one.

Furthermore, we should beware of jumping to a quick conclusion about the nature and strength of any declaratory power, without exploring how it functions in the context of the broader strategic incentives motivating political respect for judicial rulings in a specific institutional and political context. ¹⁹¹ Viewed in this broader perspective, it may well be that a particular declaratory power in one country is just as strong (if not stronger) than a strike-down power in another system. The declaration of incompatibility in the UK system is treated as binding on other branches of government and functions in a way which is not dissimilar to a strike-down power in most cases. By focusing on the formal design features of the system of judicial review, the distinction between strong-form and weak-form review tends to obscure these facts—facts which go directly to the strength or weakness of any system.

The third problem relates to an instability in the way in which Tushnet uses the terms “strong-form” and “weak-form.” When Tushnet first introduced the distinction, it was intended as a contribution to comparative constitutional scholarship of the typological variety. ¹⁹² The terms were used as a way of conceptualizing, and then distinguishing between, two systems of rights-based constitutional review (the US system on the one hand, and the UK, Canada, and New Zealand on the other). However, Tushnet went on to use the terms “strong-form” and “weak-form” to refer variously to different types of institutions, ¹⁹³ courts, ¹⁹⁴ issues, ¹⁹⁵ rights, ¹⁹⁶ statutes, ¹⁹⁷ cultures, ¹⁹⁸ or remedies. ¹⁹⁹ Thus, “strong rights” can co-exist with “weak remedies,” and vice versa. ²⁰⁰ Courts may choose to give rights a “strong or weak-form treatment ” ²⁰¹ or “ effect. ” ²⁰² At other times, Tushnet describes the phenomenon of ‘a strong-form court’ engaging in “weak-form review,” ²⁰³ or strong-form review within a weak-form culture, or “weak-form judicial review within a strong-form system.” ²⁰⁴ Within one system, he argued that there could be “strong-form review with respect to some constitutional issues, weak-form review with respect to others.” ²⁰⁵

To make matters more complicated, it sometimes seems as if Tushnet equates a deferential or restrained standard of review with weak-form review. ²⁰⁶ At others, he resists this idea because he seems to associate judicial deference or restraint with a substantive standard of review which goes to the definition or scope or ‘strength’ of the right, ²⁰⁷ whereas he associates weak-form review more closely with ‘what courts do when they find violations of the weak or strong rights, which goes to the remedies available for violations.” ²⁰⁸ “Strong remedies” include legislative invalidation, mandatory injunctions or other “coercive orders to the political branches” which have the effect of eliminating the rights-violation immediately. ²⁰⁹ Weak remedies, in contrast, tend to be more provisional and less final or intrusive. ²¹⁰ They include declaratory judgments of various kinds and deferred remedies which allow the legislature or administration time to address the rights-violation. ²¹¹ An example of the latter would be “a requirement that government officials develop plans that hold out some promise of eliminating the constitutional violation within a reasonably short but unspecified time period.” ²¹²

The problem is that once the terms “strong form” and “weak form” are used in this capacious way to refer to this wide range of phenomena which occur in all systems, then their use as a taxonomy of different systems begins to look unstable. At the very least, what began as quite a polarized picture of different systems of review, ends up being a more complex picture of multiple “blended systems,” ²¹³ all of which combine some “strong-form” and “weak-form” features in various ways and to varying degrees. ²¹⁴ Certainly, the features which Tushnet labels “weak-form” are by no means unique to the new Bills of Rights in the Commonwealth countries and are prevalent in many “strong-form systems” too. Thus, in his discussion of deferred remedies (which he characterizes as “weak form remedies”), Tushnet draws most of his examples from countries such as Canada, South Africa, India, Ireland, and the US—all of which possess a judicial strike-down power as well as other types of “strong” remedies such as mandatory injunctions. ²¹⁵

When we look at the reality of how such systems operate (including the US), the courts do not rely exclusively (or perhaps even primarily) on the strong, coercive remedies. In many systems which possess a judicial strike-down power, judges tend to use that power as a measure of last resort, precisely because it is perceived to be a drastic or blunt instrument to deal with potential rights-violations. ²¹⁶ Often, judges in such systems rely on a presumption of constitutionality to forestall findings of unconstitutionality unless it is absolutely necessary to do so. ²¹⁷ And even when judges use the strike-down power in strong-form systems, they sometimes soften its blow by suspending its effects to allow the legislature time to decide how to deal with the problem. ²¹⁸ These devices are ways of giving legislatures time, discretion and flexibility as to the most appropriate way of remedying a rights-violation. ²¹⁹ Their prevalence in strong-form systems shows that the language of “displacement” is too crude to capture the complexity of the inter-institutional dynamics surrounding questions of rights. Often, the courts in strong-form systems find subtle ways of allowing the legislature considerable room for maneuver on how to remedy rights-violations. As Tushnet acknowledges, even in strong-form systems, judges can choose whether to give rights a “strong or weak-form treatment .” ²²⁰

Apart from remedies, there are other devices and approaches which judges can use to calibrate the strength or intrusiveness of their powers, including doctrines of standing, mootness, non-justiciability, prospective overruling, as well as doctrines of deference and variable standards of review. ²²¹ Interestingly, even though Tushnet places archetypal reliance on the US as the paradigm of “strong-form review,” he takes the US experience with deferred remedies, ²²² deferential standards of view, ²²³ the political questions doctrine, ²²⁴ doctrines of “structural due process,” ²²⁵ and clear statement rules ²²⁶ as American examples of “provisional rather than conclusive review.” ²²⁷ These doctrines and devices show that even in the US, there are widespread “resonances” ²²⁸ of weak-form review.

Tushnet also acknowledges the “dialogic” or “common-law” character of some areas of US constitutional law ²²⁹ as well as the prevalence of techniques of “democratic experimentalism” ²³⁰ in the US system. Thus, Tushnet argues that the US doctrine on free speech has developed in a dialogic fashion, whereby the courts make tentative or provisional findings, wait to see how it develops or applies in practice and then adjusts its further rulings in light of the developing practice. ²³¹ Tushnet is not alone in describing this type of ongoing development of constitutional doctrine in the US as an iterative, dialogic process whereby the courts progress incrementally and tentatively, allowing the legislature and/or executive to contribute to the joint enterprise of working out the proper scope and limitations of a constitutional right. ²³² Interestingly, in his recent book on the so-called “Commonwealth model” of constitutionalism, Stephen Gardbaum pulls back from his earlier position that inter-institutional dialogue was a distinctive or unique feature of “weak-form review.” He now accepts that these forms of inter-institutional dialogue occur in both strong-form and weak-form systems. ²³³ Therefore, “dialogue” is not a distinctive marker of “weak-form systems.” Rather, it is something both systems have in common.

The array of judicial doctrines, devices and techniques which are prevalent in strong-form systems show that the metaphor of “displacement” is too crude to capture the full array of techniques available to the judiciary in any system of judicial review. These broader techniques are crucial ways in which the courts calibrate their relationship with the elected branches of government and distribute tasks between them in any system. Rather than relying on the image of one institution “displacing” another in an all-or-nothing fashion, what happens when we look in more detail at all systems of review is that judges rely on an array of much more subtle devices, techniques and remedies, many of which invite contributions from, and collaboration with, the other branches. ²³⁴ Reliance on these techniques points up the limits of a polarized distinction based on constitutional design and suggests instead the merits of viewing all systems as “blending” features of varying potency and effect in a multi-dimensional way. ²³⁵ By urging us to focus too narrowly on the availability or otherwise of the strike-down power, the distinction between strong-form and weak-form review exaggerates the differences between so-called strong-form and weak-form systems and obscures the features they have in common. Such a narrow lens distorts the true picture of not only weak-form systems but also of the way many strong-form systems operate in practice.

6. Conclusion

Katharine Young has suggested that the function of a typology is to “classify previously disjointed features and present clusters of analysis that were previously kept apart.” ²³⁶ Nonetheless, she warns that “in advancing new clusters, and the insights they deliver, typologies may create blind spots and contradictions.” ²³⁷ This article has tried to highlight some of the blind spots and contradictions which ensue from using the distinction between strong-form and weak-form review as a way of conceptualizing different systems of review. This is not to deny that examining the UK, Canada and New Zealand as instantiations of a single, over-arching “model” of constitutionalism may yield some insights. But in carrying out this examination, it is important not to distort our understanding of how these systems work in practice or to succumb to a false polarization (or indeed false commonality) between systems which the reality does not bear out. ²³⁸ Reservations about the cluster of “commonwealth constitutionalism” or weak-form review are beginning to emerge. ²³⁹ My analysis suggests that in order to understand how any of these systems work (and therefore to classify them accurately) we should engage in a much more fine-grained, contextual comparativism which takes on board the multi-dimensional nature of the interactions between courts and legislatures in all systems of rights-protection. ²⁴⁰

Two key consequences emerge from my analysis and require further study. The first is that we have to face up to problems with the idea of the “last word,” both as an empirical fact about Bills of Rights in countries like the UK, and more generally as a tool of theoretical and doctrinal analysis in constitutional law. There is no doubt that when scholars began to argue that the rights-protecting measures in the UK, Canada and New Zealand embodied a new model of “Commonwealth constitutionalism,” ²⁴¹ the model’s unique selling point was that it seemed to give the elected branches of government the “last word” on whether and/or how to protect rights. But as experience with these systems has accumulated, what has emerged is that Parliaments have not used their formal powers to override court decisions to any significant degree. ²⁴² In Canada, the notwithstanding clause has hardly ever been invoked by the Federal Government ²⁴³ and the UK Parliament has never overridden or rejected outright a judicial declaration of incompatibility. Nor has it ever overruled or modified a § 3 interpretation, even in cases where there is evidence of considerable political disgruntlement about the judicial decision. ²⁴⁴

If we join with Mark Tushnet in thinking that one of the virtues of weak-form review is that it is “supposed to impose political costs on the government by drawing public attention to the possibility that the government has violated constitutional rights,” ²⁴⁵ then we have to face the fact that legislatures (at least within the constitutional cultures prevailing in the Commonwealth countries) are generally unwilling to pay those costs. As Mark Tushnet observed, “these costs may be sufficient to convert a decision that nominally can be overridden into one that is effectively final,” ²⁴⁶ thus collapsing the distinction between strong-form and weak-form review into the theoretical possibility of exercising a power which is never exercised in practice.

Interestingly, while the strong-form/weak-form distinction urges us to focus primarily on the formal channels for disagreement between the courts and legislatures, the real constitutional action in these countries seems to be occurring at a more informal level, through reliance on a matrix of doctrines and devices which facilitate various forms of inter-institutional interaction and collaboration between the courts and the elected branches. ²⁴⁷ Thus, in the UK, one of the most important and hotly contested issues in the academic commentary has been the extent to which judges should defer (if at all) to the considered decisions of the democratic assembly and, if so, to what extent. ²⁴⁸ This kind of judicial self-restraint is just one way in which judges can calibrate their respect for the institutional competence and legitimacy of the elected branches of government, thus facilitating a division of labor between them. The centrality of this doctrine to the case-law also highlights that the shortcomings of the “last word” approach to understanding the complex inter-institutional dynamics under the HRA. Rather than thinking in absolutist terms about which branch of government has “the last word” in a constitutional dialogue, we should broaden and deepen our analysis to examine the nature and form of the various contributions made by all branches of government to protecting rights. ²⁴⁹ On this more nuanced view, both the courts and the legislature have a shared responsibility for the protection of rights, where each performs different (though complementary) roles. Thus, in the UK, it may be that the judicial word is effectively final on whether rights have been violated in particular cases which come before them, but if the courts issue a declaration of incompatibility, Parliament’s word is final on how to remedy that violation in legislative form. ²⁵⁰ This more differentiated division of constitutional labour is not easy to capture using the simple metaphor of “dialogue.”

There is one final consequence which bears mention here. This is that, at least as instantiated in the UK and Canada, “weak-form review” is not the quick fix to the legitimacy problem faced by American-style judicial review, as many scholars hoped. There is no doubt that one of the great virtues of “weak-form review” was thought to be that it solved the legitimacy problem by reducing the powers of the courts and giving legislatures the “last word” on whether and/or how legislation should protect rights. Taken at face value, weak-form review seemed to resolve (or at least mitigate) the tension between democratic self-government and rights-based constitutional review. But if legislatures do not want, or at least do not assert their power to have “the last word” to any great degree, then the legitimacy problem remains largely unabated. It is worth noting that when Jeremy Waldron argued that his objections to rights-based review only apply to strong review on the American model, he defined “strong judicial review” as a system where the courts have

the authority to decline to apply a statute in a particular case (even though that statute on its own terms plainly applies in that case) or to modify the effect of a statute to make its application conform with individual rights (in ways that the statute itself does not envisage). ²⁵¹

That definition brings the UK HRA squarely within the definition of “strong judicial review” and incurs all the legitimacy problems which Waldron levels at the US system. These problems may not be insurmountable, ²⁵² but Waldron is surely right that they must be addressed head-on as part of the justificatory challenge. ²⁵³

Finally, if legislatures do not avail of the opportunity to override court decisions in weak-form systems, then this undercuts the argument that judges should adopt an interventionist and assertive judicial posture across the board, on the assumption that their decisions will be overturned by Parliament if there is a democratic will to do so. ²⁵⁴ As we have seen, this assumption is typically unwarranted in the UK, because of the various legal and political pressures which bear on the Government and Parliament to comply with judicial decisions. If the possibility of Parliament overturning a court-ruling is actually a fiction, then it should not be used as a reason to embolden judicial spirits in weak-form systems. This article has shown that the UK courts have considerable power to protect rights in primary legislation using the tools given to them by the HRA. With great power comes great responsibility. Therefore, when carrying out their duty to protect rights under the HRA, UK judges should act with comity towards the other branches, ever mindful of the limitations of the judicial role. ²⁵⁵